Table 2. List of differentially-abundant pathway classes.

Pathway classes that significantly differentiate between three clusters	(Median) Pathway Class Abundance (at PEC-70)			Vikodak Pathway Exclusion Cut-off (PEC) threshold
	Cluster 1a	Cluster 1b	Cluster 2	50	60	70	80	90
Folding sorting and degradation	16059	17677	13281
Membrane transport	11291	12558	7452
Immune diseases	2366	3056	1393
Signaling molecules and interaction	1572	2062	981
Cell motility	4583	5383	2629
Glycan biosynthesis and metabolism	28951	37763	18875
Lipid metabolism	11971	15525	7795
Metabolism of terpenoids and polyketides	17471	22573	12049
Infectious diseases Bacterial	1178	1508	849
Metabolism of other amino acids	9603	12268	7111
Energy metabolism	46982	57159	33639
Cell growth and death	7081	8942	5820
Replication and repair	30060	36443	25571
Translation	21133	24844	18784
Transcription	2715	3342	2544
Metabolism of cofactors and vitamins	40900	50412	49879
Carbohydrate metabolism	104526	117542	151564
Nucleotide metabolism	40603	48813	67972
Amino acid metabolism	34599	38770	77688
Xenobiotics biodegradation and metabolism	3971	4869	23764
Biosynthesis of other secondary metabolites	454	543	4512
Chemical structure transformation maps	3	2	176

List of differentially abundant pathway-classes (identified between Cluster-1a, Cluster-1b and Cluster-2). Significantly different pathway classes were identified using Kruskal-Wallis rank sum test (with Benjamini-Hochberg corrected p-values < 0.001 at a False Discovery Rate of 0.0001) coupled with a bootstrap approach. Pathway-classes with significantly different median abundances in at least 99% of iterations are shown in this table. The last column titled ‘Vikodak Pathway Exclusion Cut-off (PEC) threshold’ indicates the PEC value thresholds at which the pathway-class was reported by Vikodak. Green indicates presence and red indicates absence.